Hi, I just did a 1:24 with Senna's 1994 Williams at Montreal and generally lap around 1:28. All assists off except for auto clutch and invulnerability. According to this, I beat pole by 2 seconds in 1994. How accurate is this? Reason I ask, I'm pretty much an average driver at F1 games and noob to sim life. Edited update: I just read that the track inserted a temporary chicane and an effort was made to slow the cars down that year. Would explain why both I and an 80% AI was quicker than pole.
Realize too, every car at every track is an approximation. Sim creators do not have enough data to exactly mimic any car. (some manufactures actually hide data from sims) Tires, weather, also come into play. Your best bet for trying to gauge how well a car simultates the real thing is to use Laser Scanned tracks. Unfortunately, you probably won't find too many courses that have been laser scanned in the past 3 or 4 years that match what a circuit was 40 or 50 years ago.
You could try one of the official cars at an official track and compare. For example, the McLaren MP4/8 at Soa Paulo.
Thanks for replies. Yes, I get the feeling it is easier to drive a 1994 Williams than a later car. Some of it is true due to the pace being slower but I suspect some of it is not and on other sims it is even the opposite feel. I still enjoy driving it though. Also, yes, it is hard to know what was really going on with regards to gears and traction in some F1 cars down the years, and tracks as well. I'll not pretend I can qualify within the 107% rule yet.
Here is a video of a 1:15.7 lap in a short race: In the 1993 gp, the MP4 best time was 1:17.697 (almost 2 seconds slower than the Williams' pole), while fastest lap in race was 1:20.024 (by another car). So it seems a bit faster in rF2.
Consider the rf2 version of interlagos Is wider than the real one, so the rf2 mclaren is probably more than a bit faster
If you really want to measure accurate pace, then you will have to limit yourself to laser scanned tracks. As stated above, the layout differences of any given hand drawn track could be be milometers or yards. So first you try and eliminate the physical variables. Then there is weather...was it a green track? I've read a-lot about old Can-Am cars and almost every year the new cars clearly outpaced the previous models(even F1 times when they raced at the same tracks). Yet at a few tracks, weathering of the track surface prevented the new fangled cars from besting times set the year before. Rain? Boiling Sun? I think if you sampled every sim, especially ones that try to 'be-it'all' like rF2, you will find almost all will have plenty of sim-racers who obliterate world records. Maybe a sim that is dedicated to one type of car, like ACC, that might provide closer to real world times. But that is merely a guess, I've never investigated the ACC records.
I'm helping a modding team improve the physics and handling of a sports car at the moment for an update to an existing mod. The team had already used references to real world dyno data. We chose a track where we had recent real world data and while working on things like steering ratio, spring rates and grip balance and setup we kept referencing real world lap time data. We also referenced top speeds and cornering speeds from on board videos. We started with a car that was equalling pole but didn't feel right. We ended with one that feels a lot more like real life but is 0.5 sec slower (but probably I would be 0.5 off pole). So that car on that track will be pretty accurate. However there will probably be the chance that on certain tracks the car will be faster or slower than IRL due to slightly different real world grip or torque or drag figures. I guses too that IRL a formula 1 car is an evolving thing even within one season and that's before we think about track specific upgrades. So getting it spot on can't be easy.
An important element is that top tier motorsport rocks up at a venue and then has a limited number of sessions to refine a set-up with a limited set of resources (which aren't necessarily optimised for the circuit). We can refine set-ups and configurations to our hearts' content in near perfect conditions hour after hour. And that is before we get to run more laps than our heroes ever had the opportunity to. An accurate representation of car and circuit will deliver an appreciably faster lap time in sim than was ever achieved historically.
Cars and tracks can be modelled close enough that not only are lap times close. Suspension setups can end up incredibly close to what the same driver would select on the track.
And then there is the completely unaccounted-for aspect of the fear-factor, generally quite powerful in reality, not so much in simulation. I had a friend who was said to be fearless in RL but, he died in an accident (in RL).